Using the Flocculation Index to optimise coagulant dosing during drinking water treatment. (February 2023)

• Record Type: Journal Article
• Authors: Khedher, Mahmoud; Awad, John; Donner, Erica; Drigo, Barbara; Fabris, Rolando; Harris, Martin; Braun, Kalan; Chow, Christopher W.K.
• Is Part Of: Journal of water process engineering. Volume 51(2023)
• Publication Date: 2023-02
• Subjects: Drinking water treatment -- Photometric dispersion analyser -- Turbidity -- Coagulation optimisation -- Floc characteristics; Water-supply engineering -- Periodicals; Saline water conversion -- Periodicals; Seawater -- Distillation -- Periodicals; Sanitary engineering -- Periodicals; Sewage -- Purification -- Periodicals; 627
• DOI: 10.1016/j.jwpe.2022.103394
• Languages: English
• ISSNs: 2214-7144

Abstract

Abstract: Flocs produced during drinking water treatment play a key role in particle separation and removal. Floc characteristics affect both sedimentation and filtration efficiency, however, the methods used to optimise the coagulant dose during treatment do not usually take these into account. This study investigated the feasibility of using the Flocculation Index (FI), obtained from a photometric dispersion analyser (PDA) during flocculation, as a control parameter to optimise turbidity removal. Five waters, natural and synthetic, with different turbidity levels (from 6 to ⁓120 NTU) and low DOC (i.e., <5 mg/L) were used to study the effects of various alum doses on floc characteristics [i.e., initial floc aggregation (IFA), plateau (P) and variance (VAR)]. No correlation was found between the IFA and VAR values and the alum dose achieving the greatest turbidity removal and relative settling factor (RSF). However, P followed a similar pattern to the RSF, and was negatively correlated with the residual turbidity. An optimisation model was developed using response surface methodology and the effects of two independent variables (i.e., raw turbidity and alum dose) on P values were investigated. The model achieved a high correlation (R 2 of ⁓98 %, p -value <0.05) and was validated using waters with different turbidity levels. It predicted the alum dose within 8 % error. The results indicate that the FI can be used as an alternative parameter for optimising the coagulation process. Graphical abstract: Unlabelled Image Highlights: Flocculation Index (FI) used to monitor floc properties during the treatment process Optimum alum dose for floc settling was higher than that for turbidity removal by 20 %. The plateau value (P) was found to be the key parameter for the optimisation purpose. A model to predict the alum dose based on P and turbidity levels was developed. The calibrated model was reliable to predict the optimum dose with error < 10 %.